Contralateral Clinically Unaffected Eyes of Patients With Unilateral Infectious Keratitis Demonstrate a Sympathetic Immune Response

Abstract

Purpose: To analyze the contralateral unaffected eyes of patients with microbial keratitis (MK) for any immune cell or nerve changes by laser in vivo confocal microscopy (IVCM).

Methods: A prospective study was performed on 28 patients with MK, including acute bacterial, fungal, and Acanthamoeba keratitis, as well as on their contralateral clinically unaffected eyes and on control groups, which consisted of 28 age-matched normal controls and 15 control contact lens (CL) wearers. Laser IVCM with the Heidelberg Retinal Tomograph 3/Rostock Cornea Module and Cochet-Bonnet esthesiometry of the central cornea were performed. Two masked observers assessed central corneal dendritiform cell density and subbasal corneal nerve parameters.

Results: The contralateral clinically unaffected eyes of patients with MK demonstrated significant diminishment in nerve density (15,603.8 ± 1265.2 vs. 24,102.1 ± 735.6 μm/mm²), total number of nerves (11.9 ± 1.0 vs. 24.9 ± 1.2/frame), number of branches (1.7 ± 0.2 vs. 19.9 ± 1.3/frame), and branch nerve length (5775.2 ± 757.1 vs. 12,715.4 ± 648.4 μm/mm²) (P < 0.001 for all parameters) compared to normal controls and CL wearers. Further, dendritiform cell density in the contralateral unaffected eyes was significantly increased as compared to that in controls (117.5 ± 19.9 vs. 24.2 ± 3.5 cells/mm², P < 0.001).

Conclusions: We demonstrate a subclinical involvement in the contralateral clinically unaffected eyes in patients with unilateral acute MK. In vivo confocal microscopy reveals not only a diminishment of the subbasal corneal nerves and sensation, but also an increase in dendritiform cell density in the contralateral unaffected eyes of MK patients. These findings show bilateral immune alterations in a clinically unilateral disease.

Figure 1

Corneal subbasal nerve plexus and dendritiform cells. (A) Slit-lamp photo of normal control. (B) Slit-lamp photo of control contact lens user. (C) Slit-lamp photo of contralateral clinically unaffected eye. (D) Slit-lamp photo of fungal keratitis. (E–H) In vivo confocal microscopy (IVCM) images of subbasal corneal nerve plexus in each group. (E) IVCM image from normal control. NeuronJ tracings of main nerve trunks of the subbasal nerve plexus. (F) NeuronJ tracings of total nerves, including main nerve trunks and branches of the subbasal nerve plexus in a control contact lens user. (I–L) Representative IVCM images of epithelial dendritiform cells (shown with arrows) in each group. (L) IVCM images from eye with fungal infection. Reduction of the corneal subbasal nerve plexus with increased density of the dendritiform cells is observed in the infected and the contralateral clinically unaffected eye of patients with microbial keratitis.

Figure 2

Dendritiform cell density in contralateral clinically unaffected eyes in microbial keratitis (MK) patients, their infected eye, normal controls, and control contact lens (CL) users. Contralateral unaffected eyes of patients with MK had significantly higher epithelial dendritiform cell density compared to normal controls and to control contact lens users. Error bars represent standard error from the mean. *P < 0.0001 compared to control group. ŦP < 0.001 compared to control group contact lens user. +P < 0.001 compared to infected eye. Statistical analysis performed by ANOVA with Bonferroni correction for multiple comparisons.

Figure 3

Comparison of subbasal corneal nerve alterations in contralateral clinically unaffected eyes in microbial keratitis (MK) patients, the infected eyes, normal controls, and control contact lens (CL) users. Contralateral unaffected eyes of patients with MK showed statistically significant diminishment of most of the corneal nerve parameters compared to normal controls and control contact lens users. Affected eyes, as expected, showed a bigger decrease in nerve parameters as compared to controls. (A) Total number of nerves. (B) Total nerve length. (C) Number of main nerve trunks. (D) Main nerve trunk length. (E) Number of branches. (F) Branch nerve length. Error bars represent standard error of mean. *P < 0.001 compared to normal control group. ŦP < 0.001 compared to control group contact lens (CL) user. +P < 0.001 compared to affected eye. Statistical analysis performed by ANOVA with Bonferroni correction for multiple comparisons.

Figure 4

Corneal sensation by Cochet-Bonnet esthesiometry in contralateral clinically unaffected eyes in microbial keratitis (MK) patients, infected eyes, normal controls, and control contact lens (CL) users. Contralateral unaffected eyes of patients with MK had decreased corneal sensation compared to normal controls and to control contact lens (CL) users (P = 0.64), but this was not as noteworthy as in the infected eye. Error bars represent standard error of mean. *P < 0.001 compared to normal control group and contact lens (CL) user. +P < 0.001 compared to infected eye. Statistical analysis performed by ANOVA with Bonferroni correction for multiple comparisons.

Figure 5

Correlation between dendritiform cell density increase and corneal nerve diminishment per group. A linear model between total nerve length and the log of dendritiform cell density was assumed. R² and P values are shown, indicating a significant correlation. Statistical analysis performed by Pearson correlation and linear regression analysis.